ZXing-C++

This project is a C++ port of ZXing Library.

Features

• In pure C++14, no third-party dependencies
• Stateless, thread-safe readers/generators
• Wrapper to create WinRT component
• Wrapper for Android
• Wrapper for WebAssembly

Supported Formats

Same as ZXing, following barcode are supported:

1D product	1D industrial	2D
UPC-A	Code 39	QR Code
UPC-E	Code 93	Data Matrix
EAN-8	Code 128	Aztec (beta)
EAN-13	Codabar	PDF 417 (beta)
	ITF
	RSS-14
	RSS-Expanded

Web Demos

• Scan barcodes
• Generate barcodes
• Scan with camera

WinRT Package

A nuget package is available for WinRT: huycn.zxingcpp.winrt. To install it, run the following command in the Package Manager Console

PM> Install-Package huycn.zxingcpp.winrt

Getting Started

The wrappers export very simple API to use, check BarcodeReader and BarcodeGenerator in each wrapper. For more fine-grain control in scanning process, check MultiFormatReader class. For more customization when generating particular barcode format, you need to instantiate appropriate writer, see MultiFormatWriter for more details.

Build Instructions

For Windows Desktop with VS

1. Download and install CMake if it's not already installed.
2. Open CMake GUI, specify wrappers/gdiplus as source folder in the first input, specify the build output in the second input, and click on Generate.
3. At prompt, select "Visual Studio 14 2015" (or "Visual Studio 14 2015 Win64" if you want to build for x64 platform); leave the second input (Optional toolset...) empty; leave "Use default native compilers" checked; and click on Finish to generate the VS project. At the end, you will get a solution (.sln) in your binary output directory that you can open in VS. The project ZXingGdiPlus in the solution will generate a static library.

For Windows Universal Platform

1. Download and install CMake 3.4 or more recent if it's not already installed.
2. Edit the file wrappers/winrt/BuildWinCom.bat to adjust the path to your CMake installation.
3. Double-click on the batch script to run it.
4. If the build succeeds, it will put the results in the folder UAP which is ready-to-use SDK extension.

For Android NDK

Note: The original Java-only ZXing project has a very good support for Android, whether you want to use it as external app via Intent or directly integrated into your app. You should consider using it first before trying this library since involving with native code is always more complex than Java-only code. Performance-wise, except for specific usecases, you won't notice the difference!

1. Edit wrappers/android/jni/Application.mk and adjust for your project.
2. On command line, cd into wrappers/android/jni, type ndk-build (or ndk-build -j <number of your CPU cores>)
3. Copy files in libs and java into corresponding folders of your Android project.

For web browser (WebAssembly)

1. Install Emscripten if not done already.
2. In a empty build folder, invoke cmake from emconfigure create Makefile, using wrappers/wasm/Toolchain-Emscripten.cmake as toolchain file. For example:

EMSCRIPTEN_PATH=<path to your Emscripten installation, e.g. ~/emsdk/emscripten/tag-1.38.21>
SOURCE_BASEDIR=<zxing-cpp-dir/wrappers/wasm>
emconfigure cmake -G "Unix Makefiles" -DCMAKE_BUILD_TYPE=Release -DCMAKE_TOOLCHAIN_FILE="$SOURCE_BASEDIR/Toolchain-Emscripten.cmake" -DEMSCRIPTEN_ROOT_PATH="$EMSCRIPTEN_PATH" "$SOURCE_BASEDIR"

3. Invoke emmake to create zxing.js and zxing.wasm

emmake make

4. Copy these two files to your web folder and create HTML page that includes zxing.js.

See usage example of exported functions from demos.

By default, both encoder and decoder are included. If you don't plan to use either of them, you can disable it to reduce generated code size. To do so, in the line emconfigure cmake ... above, pass -ENABLE_ENCODERS=0 to disable encoders or -ENABLE_DECODERS=0 to disable decoders.

For other platforms

Wrappers are provided as convenient way to work with native image format. You still can use the library without a wrapper.

To read barcodes:

1. Create a LuminanceSource instance. This interface abstracts an image source. You will need a third-party library to read your images. If you already have an image uncompressed in memory and you know its layout, you can easily go with GenericLuminanceSource. Otherwise, you will need to come up with your own implementation of the interface.
2. Use the LuminanceSource instance above to create an instance of BinaryBitmap. You have choices between HybridBinarizer or GlobalHistogramBinarizer. See class document in header files for more details on theses choices.
3. Create an instance of MultiFormatReader with appropriate hints. Pay attention to possibleFormats(), shouldTryHarder(), shouldTryRotate(). These parameters will affect accuracy as well as reader's speed.
4. Call MultiFormatReader::read() with the BinaryImage created above to read your barcodes.

To write barcodes:

1. Create a MultiFormatWriter instance with the format you want to generate. Set encoding and margins if needed.
2. Call encode() with text content and the image size. This returns an BitMatrix which kind of binary image of the barcode where true == visual black and false == visual white.
3. Convert the bit matrix to your native image format.